Game2Real-Bench: Do Toy Games Reveal and Transfer to Real-World Visual Reasoning? Paper • 2610.18834 • Published Oct 13 • 6
FC-SWE: Failure-Conditioned RL for Long-Horizon Software Engineering Agents Paper • 2610.07898 • Published Oct 6 • 8
GlobeBench: Evaluating Text-Based World Model Capabilities for Scalable Environment Simulations Paper • 2610.22708 • Published Oct 15 • 7
EvoWeb: A Benchmark for Coding Agents on Evolving Web Applications Paper • 2610.16402 • Published Oct 12 • 10
Librarian Agents: Training LLMs to Curate Long-Context Workspaces that Anticipate Future Queries Paper • 2610.22273 • Published Oct 15 • 9
Failure Recovery via Belief Updates and World Modeling for Long-Horizon Agents Paper • 2610.18082 • Published Oct 13 • 9
PRISM: PRIvileged Sibling-guided Multi-Agent Topology Design Paper • 2610.20628 • Published Sep 29 • 10
BRACE: Adapting Whole-Body References for Force and Terrain Aware Humanoid Motion Tracking Paper • 2610.07052 • Published Oct 5 • 9
Dream2Discern: Learning from Hard Latent Variants for Runtime Failure Prediction in Multi-Agent Systems Paper • 2610.16071 • Published Oct 12 • 10
Scene-Designer: Learning Design Intelligence for Agentic 3D Scene Generation Paper • 2610.23121 • Published Oct 16 • 13
DuoDiscover: Scaling Autonomous Discovery Beyond the Evaluation Bottleneck Paper • 2610.17184 • Published Oct 12 • 15
SciΠvot: Benchmarking Principal-Investigator Judgment at Pivotal Moments of Scientific AutoResearch Agents Paper • 2610.21082 • Published Oct 15 • 15
Towards Looped Models Done Right, Part I: Topology, Input Injection, Recurrent-State Design Paper • 2610.23036 • Published Oct 16 • 18
Lean4AgenticRL: Advancing Agentic RL with Verifiable Multi-level Process Rewards Paper • 2610.22448 • Published Oct 15 • 20
DiscoveryRSI: Recursively Learning to Discover Valuable Hypotheses Paper • 2610.23839 • Published Oct 16 • 21
OfficeTown: Generative Organizations for Evolving Agent Intelligence Paper • 2610.23194 • Published Oct 16 • 23
RAP: A Dataset for Paper-to-Poster Generation via Declarative Markup from Human-Crafted Exemplars Paper • 2610.23230 • Published Oct 16 • 28
LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets Paper • 2610.09872 • Published Oct 7 • 28
Tree-Aware Posterior Distillation for Long-Horizon Agentic Reinforcement Learning Paper • 2610.23181 • Published Oct 16 • 32
From Research Goals to Scientific Closure: Evaluating Agents Through Evidence-Supported Physics Research Paper • 2610.21767 • Published Oct 15 • 41